完全跑通1.0版本

backend/app/api/news.py

@@ -1,17 +1,16 @@
-from datetime import date, datetime
+from datetime import date
 from typing import Optional
-from fastapi import APIRouter, Depends, Query
-from sqlalchemy import select, func, distinct
+from fastapi import APIRouter, Depends, HTTPException, Query
+from sqlalchemy import select, func
 from sqlalchemy.ext.asyncio import AsyncSession
 
 from ..database import get_db
-from ..models.news import ProcessedNews, RawNews
+from ..models.news import ProcessedNews
 
 router = APIRouter()
 
 
 def _serialize(n: ProcessedNews) -> dict:
-    raw = n.raw_news
     return {
         "id": n.id,
         "title_zh": n.title_zh,
@@ -23,8 +22,8 @@ def _serialize(n: ProcessedNews) -> dict:
         "category": n.category,
         "is_featured": n.is_featured,
         "featured_rank": n.featured_rank,
-        "source_name": n.source_name or (raw.source.name if raw and raw.source else ""),
-        "source_url": n.source_url or (raw.url if raw else ""),
+        "source_name": n.source_name or "",
+        "source_url": n.source_url or "",
         "published_at": n.published_at.isoformat() if n.published_at else None,
         "processed_at": n.processed_at.isoformat() if n.processed_at else None,
     }
@@ -38,7 +37,6 @@ async def get_featured(
     target = date.fromisoformat(news_date) if news_date else date.today()
     stmt = (
         select(ProcessedNews)
-        .join(ProcessedNews.raw_news)
         .where(ProcessedNews.is_featured == True)
         .where(func.date(ProcessedNews.processed_at) == target)
         .order_by(ProcessedNews.featured_rank)
@@ -57,11 +55,7 @@ async def get_news(
     db: AsyncSession = Depends(get_db),
 ):
     target = date.fromisoformat(news_date) if news_date else date.today()
-    stmt = (
-        select(ProcessedNews)
-        .join(ProcessedNews.raw_news)
-        .where(func.date(ProcessedNews.processed_at) == target)
-    )
+    stmt = select(ProcessedNews).where(func.date(ProcessedNews.processed_at) == target)
     if category:
         stmt = stmt.where(ProcessedNews.category == category)
 
@@ -87,10 +81,9 @@ async def get_dates(db: AsyncSession = Depends(get_db)):
 
 @router.get("/{news_id}")
 async def get_news_detail(news_id: int, db: AsyncSession = Depends(get_db)):
-    stmt = select(ProcessedNews).join(ProcessedNews.raw_news).where(ProcessedNews.id == news_id)
+    stmt = select(ProcessedNews).where(ProcessedNews.id == news_id)
     result = await db.execute(stmt)
     news = result.scalar_one_or_none()
     if not news:
-        from fastapi import HTTPException
         raise HTTPException(status_code=404, detail="Not found")
     return _serialize(news)

backend/app/config.py

@@ -2,7 +2,7 @@ from pydantic_settings import BaseSettings
 
 
 class Settings(BaseSettings):
-    database_url: str = "postgresql+asyncpg://pharma:pharma123@localhost/pharma_news"
+    database_url: str = "postgresql+asyncpg://ai_news:PrDTEr6tGcyWX6G2@chenwuzhu.cn:5432/ai_news"
     admin_token: str = "change-me-admin-token"
 
     initial_llm_provider: str = "deepseek"

backend/app/crawler/rss_fetcher.py

@@ -60,14 +60,15 @@ async def fetch_rss(url: str, max_items: int = 30) -> list[dict]:
     return items
 
 
-# 默认新闻源（管理页可增删）
+# 默认新闻源（管理页可增删）— URLs 经过验证可用
 DEFAULT_SOURCES = [
     # 中文
-    {"name": "国家药监局", "url": "https://www.nmpa.gov.cn/rss/yaopinxinxi.xml", "language": "zh", "category": "药品监管"},
-    {"name": "丁香园", "url": "https://www.dxy.cn/bbs/feed.xml", "language": "zh", "category": "临床研究"},
-    {"name": "医学界", "url": "https://www.yxj.org.cn/rss.xml", "language": "zh", "category": "行业动态"},
+    {"name": "中国新闻网·健康", "url": "https://www.chinanews.com.cn/rss/health.xml", "language": "zh", "category": "行业动态"},
     # 英文
     {"name": "STAT News", "url": "https://www.statnews.com/feed/", "language": "en", "category": "临床研究"},
     {"name": "FiercePharma", "url": "https://www.fiercepharma.com/rss/xml", "language": "en", "category": "行业动态"},
-    {"name": "FDA News", "url": "https://www.fda.gov/about-fda/contact-fda/stay-informed/rss-feeds/fda-news-feed/rss.xml", "language": "en", "category": "药品监管"},
+    {"name": "FierceBiotech", "url": "https://www.fiercebiotech.com/rss/xml", "language": "en", "category": "临床研究"},
+    {"name": "FDA MedWatch", "url": "https://www.fda.gov/about-fda/contact-fda/stay-informed/rss-feeds/medwatch/rss.xml", "language": "en", "category": "药品监管"},
+    {"name": "FDA Press Releases", "url": "https://www.fda.gov/about-fda/contact-fda/stay-informed/rss-feeds/press-releases/rss.xml", "language": "en", "category": "药品监管"},
+    {"name": "Nature Medicine", "url": "https://www.nature.com/nm.rss", "language": "en", "category": "临床研究"},
 ]

backend/app/main.py

@@ -6,18 +6,38 @@ from .database import create_tables, AsyncSessionLocal
 from .scheduler import start_scheduler, shutdown_scheduler
 from .api import news, admin
 from .config import settings
-from .models.news import LLMConfig
+from .models.news import LLMConfig, NewsSource
 
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
     await create_tables()
     await seed_initial_llm_config()
+    await seed_default_sources()
     start_scheduler()
     yield
     shutdown_scheduler()
 
 
+async def seed_default_sources():
+    """Insert default news sources on first run if the table is empty."""
+    from .crawler.rss_fetcher import DEFAULT_SOURCES
+    from sqlalchemy import select
+    async with AsyncSessionLocal() as db:
+        result = await db.execute(select(NewsSource).limit(1))
+        if result.scalar_one_or_none():
+            return
+        for src in DEFAULT_SOURCES:
+            db.add(NewsSource(
+                name=src["name"],
+                url=src["url"],
+                source_type="rss",
+                language=src["language"],
+                category=src["category"],
+            ))
+        await db.commit()
+
+
 async def seed_initial_llm_config():
     """Insert default LLM config on first run if none exists."""
     from sqlalchemy import select